A protease resistant insulin like growth factor binding protein 4 as a treatment for prostate cancer

The insulin-like growth factor (IGF) pathway consists of two ligands, IGF I and IGF II, their two receptors, the IGF I receptor (IGF IR) and IGF MR along with several soluble insulin-like growth factor binding proteins (IGFBPs). The IGFBPs bind to and modify IGF activity. However, proteases such as pregnancy associated plasma protein A (PAPP-A) can cleave IGFBPs, releasing active IGF I to stimulate the IGF IR. Stimulation of the IGF IR leads to activation of the Akt or mitogen-activated protein kinase (MAPK) pathway that can lead to transformation, differentiation, proliferation and angiogenesis. Increased IGF IR expression has been reported in prostate cancer tissue relative to normal or benign tissue. Therefore, targeting the IGF pathway may prove useful in treating prostate cancer.

Tissue microarrays of prostate tumours were stained with an antibody to the IGF IR. IGF IR expression was significantly associated with Gleason grade with IGF IR expression higher in Gleason grade 3 relative to normal/benign prostatic hyperplasia (BPH) tissue and lower in Gleason grade 5 relative to Gleason grade 3 tissue. There was no significant difference in plasma IGF I levels between BPH and prostate cancer patients (Gleason score 5 or Gleason score 7). However, there were significantly higher plasma levels of the IGFBP 4 protease, PAPP-A, in Gleason score 5 patients relative to BPH patients.

PC-3M-luc2 cells are a human androgen independent cell line lacking the androgen receptor. The PC-3M-luc2 cells stably express luciferase allowing them to be imaged in v iv o using bioluminescent imaging. Expression of the IGF pathway components was assessed in these cells to ensure the IGF pathway was active. Western blot analysis showed that the PC-3M-luc2 cells expressed the IGF IR which was activated by IGF I. IGFBP 3, IGFBP 4 and PAPP-A were also expressed. PC-3M-luc2 cells were shown to proliferate in response to IGF (E3R) (an analogue of IGF I that cannot be bound by IGFBPs) by bromodeoxyuridine (BrdU) assay. However, the 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) assay showed PC-3M-luc2 cells proliferated in response to IGF I. IGF (E3R) also significantly increased vascular endothelial growth factor (VEGFies) expression in PC-3M-luc2 cells.

We assessed a PAPP-A resistant mouse IGFBP 4 (dBP4) as a strategy to neutralise the IGF I ligand. dBP4 was developed by mutation of the PAPP-A cleavage site from 119-KHMAKVRDIRSKMK-133 to 119-AAMAAVADASAMA-133 and cloned into pTriEx4 Neoplasmid with C-terminal His-tag for expression in human embryonic kidney cells. Purified dBP4 was resistant to PAPP-A cleavage and retained IGF I binding capacity. In v itro , dBP4 significantly inhibited tubule formation by endothelial cells relative to controls or IGF I treated cells suggesting that dBP4 possesses anti-angiogenic properties.

A Tet inducible expression system was used in v iv o to test the effect of dBP4 on tumour growth and metastasis in subcutaneous, orthotopic and experimental metastasis models of prostate cancer. The inducible Tet-On system allowed dBP4 expression to be induced by doxycycline. Following implantation of PC-3M-luc2 dBP4 cells into non-obese diabetic, severe combined immunodeficiency, gamma 2 receptor null (NSG) mice, tumours and metastasis were monitored by bioluminescent imaging and/or caliper measurements. In the subcutaneous tumour model, dBP4-expressing subcutaneous tumours were significantly smaller than controls and in the orthotopic model, bioluminescence was reduced relative to controls. Mice bearing dBP4-expressing subcutaneous tumours survived significantly longer than controls. There was no survival difference between mice bearing dBP4-expressing prostate tumours and controls. IGF I expression within the dBP4-expressing subcutaneous and prostate tumours was decreased relative to controls. Western blot analysis of tumour lysates showed that dBP4 inhibited IGF IR activation in both the subcutaneous and prostate tumours, which may be a consequence of reduced IGF I within the dBP4-expressing tumours. However, activated Akt was only decreased in the subcutaneous tumours relative to controls. As tumours require vasculature in order to supply oxygen and nutrients essential for tumour growth, expression of VEGF was assessed in the subcutaneous tumours. Control tumours expressed VEGF164/165, which was not expressed in the dBP4-expressing subcutaneous tumours.

dBP4 expression inhibited tumour growth in both subcutaneous and orthotopic-prostate cancer models. As dBP4 expression inhibited endothelial cell tubule formation and abolished VEGF_164/165 expression within PC-3M-luc2 tumours growing subcutaneously, this

suggests that dBP4 has either anti-angiogenic or anti-proliferative properties and may

therefore have value as a therapeutic for prostate cancer.